Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The silence stretched, broken only by the hum of the stalled elevator. I watched my reflection in the polished steel doors, a stranger staring back. My eyes were dull, the lines etched around my mouth deeper than usual. The trapped feeling was not new; it was the echo of a thousand locked doors in my past.

"Well, this is just great," Michael, my coworker, said. His usual boisterous energy seemed subdued, a testament to the oppressive atmosphere. He leaned against the wall, a sigh escaping his lips.

I remained rooted to my spot, my shoulders hunched. The weight of the world pressed down, crushing. I wanted to scream, to lash out, to break something. But the energy to do so was gone, replaced by a hollow void.

He started telling a joke, a weak attempt to break the tension. I gave a small, involuntary twitch of a smile. I looked away. The thought of three hours felt like an eternity.

Emotion: depressed

Cluster: Sadness / Despair
PC1 (Valence): -2.06 Negative
PC2 (Disposition): -1.08

Role in Research

This story is one of 1,000 stories generated for the emotion depressed. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 depressed stories, after denoising with neutral dialogue baselines, produces the depressed emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the depressed vector is projected through the unembedding matrix.

Promoted:
😞0.417
每天0.353
无法0.330
own0.311
😔0.309
Suppressed:
de-0.492
B-0.393
P-0.352
l-0.310
!-0.307